It is very difficult for robot to quickly obtain surrounding environment information and establish realtime environment map to achieve autonomous operation in unknown environment. To solve the problem

present a method based on visual navigation

which uses panoramic camera as vision sensor system of robot to collect environmental information and use fuzzy clustering image segmentation to division color map in HSI space

and obtain binary image; grid processing on image is carried out

build environment map

and use 8 directions connected Dijkstra for global path planning and calculate the optimal path

so as to achieve fast and autonomous movement of mobile robot. The simulation experiments show that the method is effective and feasible.